Carmen,
I don't know of any efficiency threshold, but as a comparison you
might want to calculate the efficiency of a standard block design with
16s - 20s blocks for the same length of time as your experiment. Block
designs maximize detection efficiency, and this would tell you how
efficient your experiment is compared to that standard.
The way to make up for an inefficient design is of course to scan
longer and get more data, and that works fine up to a point, and then
subject fatigue and habituation become problematic.
Another way to see if you are on the right track is to find similar
experiments in published papers, and see how long they scanned their
subjects, and how the data turned out.
Jim Lee
On 1/15/14, Carmen <[log in to unmask]> wrote:
> Dear SPM-experts,
>
> I currently try an efficiency calculation using this equation: (d, n) =
> 1/(c'*pinv(X'*X)*c).
> For the optimal design of stimulus sequence with regard to psychological
> aspects, the efficiency for the differential contrast I am interested in,
> reaches only 0.16.
> I wondered if there is some threshold, which the efficiency should exceed?
> Thank you for your help,
> carmen
>
|